From 815e90c2278c522c62d249d0403a0154942927c8 Mon Sep 17 00:00:00 2001 From: Yoshitomo Matsubara Date: Mon, 13 Nov 2023 23:02:36 -0800 Subject: [PATCH] Update image links --- README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index ade740ef..011b0ade 100644 --- a/README.md +++ b/README.md @@ -13,10 +13,10 @@ Then, intermediate, informative features are transmitted through a wireless comm edge server that processes the bulk part of the computation (the second sequence of the model). -![R-D (rate-distortion), ExR-D, and Three-way tradeoffs](imgs/ilsvrc2012-overview.png) +![R-D (rate-distortion), ExR-D, and Three-way tradeoffs](https://raw.githubusercontent.com/yoshitomo-matsubara/sc2-benchmark/main/imgs/ilsvrc2012-overview.png) ImageNet (ILSVRC 2012): R-D (rate-distortion), ExR-D, and three-way tradeoffs for input compression and supervised compression with ResNet-50 as a reference model -![Input compression vs. Supervised compression](imgs/input_vs_supervised_compression.png) +![Input compression vs. Supervised compression](https://raw.githubusercontent.com/yoshitomo-matsubara/sc2-benchmark/main/imgs/input_vs_supervised_compression.png) Input compression is an approach to save transmitted data, but it leads to transmitting information irrelevant to the supervised task.